A major obstacle to understanding the neural mechanisms underlying behavior is our inability to distinguish between neuronal classes in recordings from behaving animals. Mouse transgenic lines, including CRE lines, have made it possible to use optogenetic techniques to selectively activate different classes to determine their role in behavior. However, it remains a challenge to determine the neuronal identity based on optogenetic stimulation, as the latency of response to direct stimulation can overlap with strong but indirect disynaptic excitation. The proposed research will introduce novel methods for distinguishing direct from indirect stimulation in order to label neuronal identity. It will furthe develop these methods for extracellular recordings made with linear array electrodes and using current source density (CSD) analysis to identify laminar location. These methods will be validated in CRE mice, and then applied to the marmoset (Callithrix jacchus). The marmoset is a particularly interesting primate model because it offers opportunities for dissecting neural circuitry that are comparable to the mouse. It matures quickly and breeds well in captivity, so it is amenable to the kinds of genetic manipulation used in mice and has produced the first primate transgenic lines. It also has a lissencephalic (flat) cortex which aids in laminar recordings. Preliminary data show marmosets can perform visual tasks under head-restraint, making them suitable to awake neurophysiology. Establishing these methods will create opportunities to study cortical circuits at a mechanistic level, enabling the field to understand how aberrations in the cortical circuitry give rise to devastating disorders such as schizophrenia and autism.